Have you noticed that you’re getting a lot less spam on Twitter these days? You may have to thank a bot for that. Twitter has just shed light on BotMaker, a recently developed system that (as the name suggests) lets the social network create anti-spam bot code with very little effort. Within a few seconds, engineers can set up rules that automatically take down and track spammers, in some cases before they’ve even managed to post anything. Besides barring known spam links, the bots can flag suspicious behavior — if a lot of people block an account after it sends a tweet, it’s going to be watched very closely. BotMaker will also look at long-term behavior, so spammers that slip through the cracks aren’t necessarily safe.
Importantly, you shouldn’t notice that BotMaker is working; it’s designed to only fight certain forms of spam as they arrive, and saves more time-consuming tasks for later. Whatever its impact on performance, it’s proving to be effective. Twitter is reporting a 40 percent drop in spam since its new tool kicked in, and it can improve any less-than-perfect rules within seconds instead of hours. It’s doubtful the technology will ever completely rid the world of pitches for fake followers and cheap drugs, but it should help staff police a rapidly growing user base without hurting your day-to-day experience.
Filed under: Internet
Comments
Source: Twitter Engineering Blog